Oxidation kinetics of biodiesel by non-isothermal pressurized-differential scanning calorimetry

Authors: Dunn, Robert - Bob

Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/5/2020
Publication Date: 7/1/2021
Citation: Dunn, R.O. 2021. Oxidation kinetics of biodiesel by non-isothermal pressurized-differential scanning calorimetry. Transactions of the ASABE. 63(3):687-701. https://doi.org/10.13031/trans.13708.
DOI: https://doi.org/10.13031/trans.13708

Interpretive Summary: Methods were developed to predict the shelf-life of biodiesel when it is stored in tanks before being used in the United States transportation sector. During long-term storage, contact with ambient air can lead to oxidation which can compromise the fuel quality of biodiesel. This work examined the reaction kinetics involved in the oxidation of biodiesel made from canola, palm and soybean oils (CaME, PME and SME). Kinetic models were fitted to data obtained by analyzing the oxidation of biodiesel under pressurized air and high temperatures. Once oxidation induction times were tabulated, they were used in a mathematical equation that predicts the shelf-life at lower temperatures than might be expected in common storage conditions. Results were correlated to oxidative stability data obtained from the standard test method recommended in biodiesel fuel specifications. This research will benefit industry, fuel producers, terminal operators and users that need to store biodiesel during warm weather for long periods of time.

Technical Abstract: Biodiesel is a renewable fuel that can be made from transesterification of plant oils, waste greases, animal fats, or microalgal oils with methanol. During long-term storage, autooxidation can adversely affect the viscosity and ignition quality of biodiesel. The objective of this work was to investigate the kinetics of oxidation of fatty acid methyl esters (FAME) made from canola, palm, and soybean oils (CaME, PME, and SME) plus methyl oleate and stearate (MeC18:1 and MeC18:2). The FAME were analyzed by non-isothermal pressurized-differential scanning calorimetry (PDSC) in dynamic mode (positive airflow) at heating scan rates of 2, 5, 10, 15, and 20 °C/min. Results were analyzed to infer oxidation rate constants (k[T]). First order and two autocatalytic kinetic models were applied to calculate isothermal oxidation induction times (OIT) for each FAME. While OIT values varied between the models, ranking the FAME in descending order of OIT was the same for each model regardless of temperature. In contrast, ranking the FAME by the oxidation induction period (IP) obtained from a Rancimat instrument depended greatly on temperature. Shelf-life data for each FAME were calculated by extrapolating IP measured at high temperatures to 25°C (SL1). Ranking the FAME by SL1 yielded trends that were unpredictable based on the IP data. On the other hand, OIT values calculated at 25°C (SL2) from the PDSC results were more consistent. These findings suggest that estimating the shelf-life of biodiesel from dynamic mode non-isothermal PDSC data yielded more reliable results than extrapolation of IP data measured at high temperatures.